For a long time we have focused much attention on traceability. Recently, of course, traceability was the buzzword that the government and non-governmental organizations focused on as a weak link in the food safety system during the Salmonella Saintpaul outbreak.
The focus on traceability as our Achilles heal of the industry was not really accurate. In fact, as our interview with Jim Gorny of UC Davis predicted, the problem was actually epidemiology, not traceability.
Yet Bruce Peterson was correct when he pointed out that the produce industry is unlikely to eliminate all food safety problems and that other industries do a better job of limiting the damage from food safety problems by having excellent traceability so that virtually all food safety problems are tied to a specific product, brand, factory, etc., simultaneous with the announcement of the problem.
Still we are not sure that the industry’s joint initiative on traceability, even if fully successful, is ever likely to achieve Bruce’s vision. After all, the true fulfillment of that vision depends heavily on consumers being able to identify what item they bought, and that depends crucially on item-level traceability and branding on that item level so the consumers can instantaneously say they purchased a clamshell of XX brand of an item and still have the clamshell in the refrigerator.
Our own take is that if the industry is really serious about traceability for food safety purposes, it leads inevitably to an embrace of a mostly packaged produce department similar to what is common in the UK and at Tesco’s Fresh & Easy stores here in the US or a requirement that individual items be stickered with GTIN Bar codes plus a lot number that embeds a pack/harvest date.
But the joint task force punted on these very expensive and deeply controversial ideas and elected to leave item level traceability for another day and, instead, devote its efforts to the more manageable idea of traceability on the case level.
Although subgroups on Pallet Labeling, Synchronization Options, Retail Barriers, Supplier Barriers and Communications will continue to meet — now being managed by PMA, United and CPMA — the main task force is finished, with these being its final words: “Having addressed the primary goals of this Produce Traceability Initiative, there was consensus that this was the last meeting of this Steering Committee.”
What is going on now is that the Produce Traceability Final Action Plan is being shopped around the industry by senior officers of the trade associations to get endorsements, and the plan is to set up a website and publicize the names of these endorsers as part of an industry solution to the traceability problem in much the way that the California Leafy Greens Marketing Agreement was trumpeted as the industry’s response to the spinach crisis of 2006.
From what we understand, the associations will have little trouble getting many endorsements to announce. However, although some of these companies are sincere about making this happen, many have told us that they felt the pressure to endorse to be unbearable — not so much from the associations but from the mere fact that a public announcement will be made. Even if they are sincere, many of the requirements under the plan, especially for buyers, are not required for several years. Whether money will ever be budgeted to implement the plan is uncertain even within the most well-intentioned of organizations.
As one important executive explained the situation to us:“Companies will want to seem progressive and sign on with no intent to follow through. Where’s the integrity in that?”
There seems to be four basic problems that leaders in the industry are having with this initiative:
1. Even if fully implemented by big players, it won’t solve the industry traceability problem. One key player explained the problem this way:
Putting in a system to trace product gets more difficult the further down we go in the distribution chain. Stand on the floor on a busy Terminal Market and try and imagine where the product goes after it is sold by the Wholesaler. A customer known as “Ken, the guy with Red truck” pays cash for a pallet of tomatoes. He takes the tomatoes to his garage where the boxes sit on the floor next to cleaning supplies, motor oil, and who know what else.
He and his kids (2 of whom just used the toilet without washing their hands) dump the tomatoes on a dirty tarp to sort them for color. The green ones sit in the garage for a few days to color up during which time one or two rodents snack on tomatoes. When they finally ripen, Ken delivers the tomatoes to some of the finest restaurants in town for all of us to enjoy.
Somehow I don’t think that Ken or even a legitimate small wholesaler or purveyor is interested in investing in a traceability system. They will have to be dragged kicking and screaming to the table. The problem is that the system is only as good as its weakest link, and unless Ken is a part of the system it doesn’t work.
It is noticeable that during the Salmonella Saintpaul outbreak, the focus seemed to be on mostly small Mexican restaurants. These are the types of venues that probably do not get their produce from Sysco. They probably get it from small purveyors who are completely outside the scope of this traceability plan. This is not a fact that will escape notice when the plan is announced.
In this sense, the problem with the plan is that it is really focused on improving traceability among the sector of the industry that has the best traceability right now, but it leaves untouched a secondary sector of the trade that may well be the real problem.
2. Foodservice seems to find the plan almost completely unworkable. Part of the concern is the practicality of the plan:
Whereas retail moves product mostly by the pallet, in foodservice, product moves often by the case. Between the investment in systems and a big productivity hit, the cost for foodservice distributors will be many millions. Who has the money for this right now?
There also is a concern about the effectiveness of the plan:
A significant part of certain produce items in foodservice distributor’s sales are in splits or breaks (packages broken down into smaller units).
Items, such as peppers, cucumbers and squash, are often sold to restaurants in units other than the shipping point container. Many major distributors use wholesalers to do those splits for them. These repacked units are typically labeled with lot numbers for traceability purposes.
Many other distributors, however, do their own splits and look at their repack area as an important profit center. In addition, many repack and color sort their own tomatoes rather than use a re-packer. The vast majority of those that do their own repacking and splits have (or at least had before the Salmonella Saintpaul outbreak made traceability a big issue) absolutely no idea what product went into the box or to which customer that box was shipped.
There also is concern about both the practicality and usefulness of the system. From a financial perspective it is important to remember that buyers pass the costs of recalls onto the suppliers. This is true of both retailers and foodservice distributors but retailers, with the reputational risks inherent in having consumer brand names, are likely to sign up because the downside of being identified as anti-traceability is too great. Foodservice distributors may be able to speak a little more bluntly — there is no ROI on this for a buyer because buyers pass on these recall costs. From a usefulness perspective, it is not clear that super specific information is all that helpful. True as we discussed here with a follow-up Mailbag one of the problems in foodservice is that recalls tend to be excessively broad. If a shipper sends in four pallets, say 320 cases of a product, with, say, three different lot numbers, because right now most foodservice distributors do not track the lot numbers, if there was a recall of one lot, the foodservice distributor has to recall all 320 cases — not just the one lot that is implicated.
Now, the ability to narrow a recall seems very important but, first, it doesn’t save the distributor any money because the distributor bills back the shipper. Second, many, maybe most, recalls are intentionally broadened by distributors out of an abundance of caution. Many would say that if a shipper sent in product on the same truck with recalled product, better to throw it all out.
In fact when we covered the Honduran cantaloupe situation, one thing we discovered was that many retailers and foodservice distributors not only discarded the implicated cantaloupes, they threw away all melons of all brands from the country of Honduras! Many executives judge it impractical and too risky to ask employees who work in produce warehouses, at retail or in the kitchens of restaurants to make these subtle distinctions and pull certain lot numbers off the shelves. They think it safer to just throw out everything. Of course if we are going to do that, the need for an elaborate traceability system is less clear.
We are told that foodservice was the most vocal at the meetings about the plan being problematic, but we are also told that this was simply a function of grower/shippers being hesitant to seem to be objecting in front of their big retail customers and big retailers hesitant to identify themselves with being hesitant to invest in traceability.
3. Many grower/shippers feel that they have been put in a ridiculous position. The plan includes timelines but they are different for different classes of trade. As one grower/shipper explained:
By the third quarter of 2010 we have to ‘Show human readable information on the case’ and ‘be ready’ to ‘Encode Information on a Barcode on the case’ but inbound scanning (by buyers) isn’t until 2011 (no quarter mentioned so the deadline is Dec 31, 2011) and outbound (once again by buyers) in 2012 (once again no quarter mentioned so the deadline is December 31, 2012). True we just have to ‘be ready’ to barcode but, even so. Why should we have to spend all that money to ‘be ready’ by third quarter of 2010 and put information on cases for over two years — with all this information not being collected or of any use? I say either we do this together, or we don’t do it at all.
Because the timelines call for grower/shippers to “go first,” many are skeptical that buyers will ever actually make the investment to use the data that grower/shippers are supposed to invest in producing. It is a chicken-and-the-egg situation — Why should shippers incur the cost and make the investment if the information is not captured further down? And shippers think, legitimately, that the end result of the traceability initiative was a decision that even if it didn’t address whether the chicken or the egg came first, the decision was clearly that between the producer and the buyer, the producer came first.
Beyond this, grower/shippers already see a one-upmanship going among retailers to demand even stricter standards. One major retailer has sent around a letter demanding full GTIN compliance on every carton by the end of 2008 — although the industry traceability plan leaves that requirement until the third quarter of 2010.
Some grower/shippers advised that the industry may be wildly underestimating the cost of the initiative. For example, a number told us that they had facilities they utilized nearby major retail distribution centers. These shippers bring product into these facilities, inspect it, if necessary do some repacking, etc. — the goal is to make sure that they meet the metrics of their customers which often require virtually zero rejections. It seemed to them as if this system required them to scan both in and out of these facilities and also required them to create a new GTIN for repacked items.
Others pointed out that if a pallet contained more than one lot, it would have to be treated as a mixed pallet, broken down and scanned box by box. But how will receivers know if it has more than one lot? Since the communication process on single item pallets would be difficult, it seems likely that receivers will simply demand that all pallets of a single product come from the same lot.
Shippers who do a lot of low volume items felt that costs would surely increase, as on mixed pallets every pallet must be broken down and each case scanned individually at each stop on the product journey.
More than a few grower/shippers already do a lot of this on the pallet level and they question, especially at retail, the sense of going to a case system. Those who sell bulk commodities often point out that they sell in pallets, and the consumer buys by the item, so tracing the carton doesn’t really accomplish much although it will be a very expensive change in the system.
4. Are buyers actually going to do this? Foodservice is clearly crying that it will be very difficult, and many grower/shippers have severe doubts about retailers. We heard plenty of doubts from plenty of retailers about the certainty that their own organizations will make the investments required
The core of the initiative is a seven-point plan. The first three, which basically relate to a decision to move to GTIN (a sort of UPC code for cases), were completely uncontroversial. Points four and five speak to packer/shipper obligations to use the GTIN both in a bar code and human readable format and seem widely accepted by the big grower/shippers — if, in fact, buyers can use the information and, crucially, if buyers are willing to constrain their supply chains and only purchase from vendors who meet these traceability requirements. Besides the truth is that points one to five are easy for big buyers to dictate if they choose to do so.
The big doubts in the program are points six and seven. These basically require buyers — both direct from the packer and any subsequent handlers — to spend a lot of money, collectively probably in excess of a hundred million dollars to both read the barcodes and electronically store the information on every inbound case and then do it again on every outbound case.
Obviously there is no direct return on investment to doing this, so many, including some retailers, doubt that buyers will spend the money unless the government comes along and orders it.
Now Tom Stenzel of United Fresh spoke at the last meeting of the task force and he pointed out that the recent food safety outbreaks have focused attention on traceability. Although he supported, of course, the industry crafting its own solution and getting ahead of the issue, he also indicated that it was quite possible that government would decide to regulate anyway in order to cover those who were not voluntarily implementing effective traceability systems.
And, in fact, the industry executives who are enthusiastic about the traceability initiative typically see the prospect of government action as a powerful driver. When asked if he thought retailers would spend the money to make this plan happen, one executive expressed his thoughts this way:
I’m guardedly optimistic, and here’s why: the Government.
There is already movement inside Washington to do “something” about the produce industry and its perceived lack of traceability.
I think the Action Plan that we now have is realistic and doable and gives the industry a very real alternative to a government-directed approach. Now with all of that said, there are some very real challenges. Foodservice and split cases are an example. And there will always be those who come late to the party. But I do believe that major receivers will, in fact, get on board. Time frames may not be precise, but moving to GTINs and GS1 bar-coding that has the GTIN plus lot number as the case identification standard makes great sense.
There is still flexibility to provide for inventory control models and other nomenclature between buys and sellers. But I think the paradigm shift in the industry is recognition that from the perspective of traceability, standards are critical. And all of the major players on both the buy side and sell side are bought in.
I recognize that “talk is cheap”, so we’ll see.
Indeed we will. Our take on all this is that the basic design of the industry traceability system is a very good. The “paradigm shift” that recognizes that traceability is best obtained with industry-wide standards rather than proprietary systems is a breakthrough.
Steve Lutz, of the Perishables Group, who facilitated the discussions, and Cathy Green, of Food Lion, who chaired the initiative, deserve much praise as do all the participants who worked hard to gain consensus.
The question now is: What is the purpose that the industry hopes to serve by making a big announcement?
It is fair to say that the California Leafy Greens Marketing Agreement was, to a substantial extent, aimed at heading off government regulation. Yet we are not certain that this is an appropriate or desirable goal in this case. After all, what made the CLGMA a credible alternative to regulation was two things: First that virtually all the production of California leafy greens was covered and second, those who signed up subjected themselves to real and meaningful standards, including a willingness to be audited by government auditors.
Neither of those conditions will exist in this case. Although the industry may be tempted to say that endorsers account for 90% of produce produced or imported into the country — that would be not exactly correct. Traceability depends on the weakest link, and one non-participating wholesaler or purveyor can break the chain — even if that case is “counted” elsewhere in the supply chain.
And even if buyers install all the systems, nothing in the endorsement of the plan requires those buyers to constrain their supply chain and only buy from producers who meet the standards. We’ve written about the double standard on food safety and locally grown produce many times including, recently, here and here.
By asking for endorsements but without any requirement that buyers agree to restrain their supply chain… much less that they agree to be audited for that fact… this traceability initiative looks like another opportunity for a big double standard. Just as many buyers use supposed enthusiasm over locally grown product as an excuse to buy from less expensive sources, surely many buyers will find reason to waive any traceability requirements and buy from the cheapest source, even if the buyers have scanners and systems to store the data.
In the battle to establish the California Leafy Greens Marketing Agreement, it was widely recognized that a mere ‘endorsement’ of the agreement by retailers was insufficient. What was required was a commitment by buyers to constrain their supply chain and only buy from signatories to the agreement. In the absence of this commitment, which is not even being requested, buyer ‘endorsement’ of the plan will ring hollow.
If this is presented as an industry solution to traceability, precluding the need for government regulation, the industry will come across as less than wholly truthful.
And, if big players are going to invest to make this happen, it is not clear why the industry should even look to head off government regulation. After all, the practical effect of doing so is to prevent sectors of the industry that have refused to make these investments to continue to operate.
Perhaps the best way to position this is not as an alternative to government regulation but, rather, as a test model for what should government legislate in this area.
Many of the objections really are about cost, and the fear on cost is not just that there is a substantial out-of-pocket expense but that the standard won’t be upheld and that it will not be a level playing field. Think about a chain such as Supervalu’s Sav-a-Lot. These stores are independently owned. Maybe Supervalu will invest in these traceability systems and maybe it will only buy from shippers who follow the rules — but who says the store owners won’t buy outside the system because someone who hasn’t invested in all this is cheaper?
If there was a government regulation here — if produce couldn’t be sold that didn’t conform to these standards or couldn’t be handled by people not capable of conforming — that would eliminate the fear of an unequal playing field and remove the possibility that buyers could subvert the system by buying cheap product from producers who didn’t invest in meeting these standards.
We should trumpet the endorsers as “early adopters” willing to pave the way for the whole industry as the proposed model is tested out by real players in real time. But in the trade’s public pronouncements we should be very modest about the initiative. It will only be implemented by big players and it will be implemented slowly. Whole sectors of the industry will struggle with it for a long time. It will do little if anything to help traceback at a small independent Mexican restaurant supplied by a local purveyor who buys off a local wholesaler who buys from a regional terminal market. It is only phase one of the response that will ultimately be required if the industry is to have a comprehensive traceability system.
Still, the accomplishment of this task force should not be underestimated. It has made the industry wrestle with important issues and the trade has come out of the process with a consensus that industry standards are a better way to go than proprietary models. The industry also has agreed on what those standards should be: GS1 bar-coding that has the GTIN plus lot number. The whole process has caused many in the industry to think and act in a new and more progressive way. That makes it a big win. It doesn’t make it a complete solution, and we pretend otherwise at the peril of the industry’s credibility.